We have observed that CD34+ progenitors from CML patients have an impaired capacity to differentiate into NK cells, proliferative poorly and have decreased cloning frequency. We hypothesize that this abnormal behavior may be a direct consequence of the BCR/ABL gene product exerting adverse, intrinsic effects on NK cell progenitors. Alternatively, products of BCR/ABL+ myeloid cells may exert adverse, extrinsic effects on NK cells and their progenitors. In either case, impairment of NK cells may contribute to the inconsistent response to immunotherapy that we have observed in clinical trials. In either case, impairment of NK cells may contribute to the inconsistent response to immunotherapy that we have observed in clinical trials. During the current finding period, we developed in vitro, stroma-dependent assays to characterize steps involved in NK cell differentiation from undifferentiated CD34+/Lin- /CD38- hematopoietic progenitors as well as from progenitors already committed to lymphoid differentiation. In SA1, we will capitalize on these assays to determine intrinsic and extrinsic effects of BCR/ABL on NK cell progenitor differentiation. BCR/ABL will be introduced into normal NK cell progenitors, and effects on differentiation will be determined. In additional experiments, we will attempt to block the effects of BCR/ABL in NK cell progenitors with antisense strategies and inhibitors of tyrosine kinase activity. In SA2, we have observed that despite an impaired capacity to differentiated and proliferate, a population of BCR/ABL+NK cells can be identified which coexists with BCR/ABL- NK cells in the blood of CML patients. We hypothesize that these BCR/ABL+ NK cells have defective function and shortened survival that may be of biologic significance in the course of CML and will test this hypothesis. In SA3, we will determine whether HLA class I on CML cells may protect them from attack by autologous NK cells by interacting with inhibitory receptors. We hypothesize that NK cells obtained from normal allogeneic donors may be better effector cells to provide tumor suppression following autologous transplant therapy for CML. The interaction between allogeneic NK cells and hematopoietic cells from CML patients will be explored. Killer immunoglobulin-like receptor (KIR) and lectin receptor acquisition and function will be assessed in CML. Autologous and allogeneic NK cell clines will be chosen for HLA class I inhibitory receptor non-identity, expanded and tested for their capacity to specifically suppress BCR/ABL+ CML myeloid clones in hematopoietic progenitor co-culture assays. In SA4, results of these studies will be used to design clinical trials in which allogeneic NK cells, or modified autologous NK cells can be used as an adjunct to autologous transplant therapy for CML.